ARTICLE IN PRESS. Journal of Biomechanics

Transcription

1 Journal of Biomechanics 43 (21) Contents lists available at ScienceDirect Journal of Biomechanics journal homepage: The influence of sagittal center of pressure offset on gait kinematics and kinetics Amir Haim a,b,n, Nimrod Rozen c, Alon Wolf a a Biorobotics and Biomechanics Lab (BRML), Faculty of Mechanical Engineering, Technion-Israel Institute of Technology, Haifa, Israel b Department of Orthopedic Surgery B, Sourasky Medical Center, Tel Aviv, Israel c Department of Orthopaedic Surgery, Ha Emek Medical Center, Afula, Israel article info Article history: Accepted 28 October 29 Keywords: Center of pressure Coronal kinetics of the knee Footwear-generated biomechanical manipulations Gait analysis Knee flexion torque abstract Objectives: Kinetic patterns of the lower extremity joints have been shown to be influenced by modification of the location of the center of pressure (CoP) of the foot. The accepted theory is that a shifted location of the CoP alters the distance between the ground reaction force and the center of the joint, thereby modifying torques during gait. Various footwear designs have been reported to significantly alter the magnitude of sagittal joint torques during gait. However, the relationship between the CoP and the kinetic patterns in the sagittal plane has not been examined. The aim of this study was to evaluate the association between the sagittal location of the CoP and gait patterns during gait in healthy men. Methods: A foot-worn biomechanical device which allows controlled manipulation of the CoP location was utilized. Fourteen healthy men underwent successive gait analysis with the device set to convey three different sagittal locations of the CoP: neutral, anterior offset and posterior offset. Results: CoP translation in the sagittal plane (i.e., from posterior to anterior) significantly related with an ankle dorsiflexion torque and a knee extension torque shift throughout the stance phase. Likewise, an anterior translation of the CoP significantly reduced the extension torque at the hip during preswing. Conclusions: The study results confirm a direct correlation between sagittal offset of the CoP and the magnitude of joint torques throughout the lower extremity. & 29 Elsevier Ltd. All rights reserved. 1. Introduction During the stance phase of the gait cycle, a force is applied to the ground which is coupled with a ground reaction force (GRF). The magnitude of the GRF is equal and its direction is opposite to the force the body exerts (Winter, 1984). Consequently, joint torques develop which are equivalent to the magnitude of the GRF and the perpendicular distance from the joint center to the force (Gronley and Perry, 1984; Winter, 1984). Theoretically, altering the instantaneous center of pressure (CoP) of the foot would influence the orientation of this force and the resulting joint torques and angles through the body segments. This principle has been the focus of previous research which examined the utilization of footwear-derived biomechanical manipulation. Application of wedge insoles were found to shift the location of the CoP in the coronal plane, thereby altering joint n Corresponding author at: Biorobotics and Biomechanics Lab (BRML), Faculty of Mechanical Engineering, Technion-Israel Institute of Technology, Haifa 32, Israel. Tel.: addresses: amirhaim@gmail.com, alonw@tx.technion.ac.il (A. Haim). torques from the foot proximally (Kakihana et al., 2; Maly et al., 22; Xu et al., 1999) and decreasing the load at the medial compartment of the knee joint in healthy and arthritic subjects (Crenshaw et al., 2; Kakihana et al., 2; Ogata et al., 1997; Yasuda and Sasaki, 1987). In a previous study (Haim et al., 28), we examined the effect of controlled coronal plane CoP modulation at the foot. The magnitude of the knee adduction torque was found to significantly correlate with the coronal orientation of the CoP. Several studies have investigated the effect of sagittal plane footwear modifications on kinematic and kinetic parameters of the lower extremities. Walking with different heel-height shoes has been reported to decrease stride length (de Lateur et al., 1991), to alter joint torques in the lower extremity (Snow and Williams, 1994), and to prolong midstance knee flexor torques during gait (Kerrigan et al., 2). Missing-heel shoes were found to reduce walking speed and stride length, to increase cadence, and to considerably alter normal ankle joint function (Attinger- Benz et al., 1998). Gait analysis of negative heel rocker sole shoes showed an increase in cadence and a significant alteration of proximal joint metrics (Myers et al., 26). Similarly, changes in CoP locus were reported with relation to rocker sole /$ - see front matter & 29 Elsevier Ltd. All rights reserved. doi:1.116/j.jbiomech

2 97 A. Haim et al. / Journal of Biomechanics 43 (21) shoes (Xu et al., 1999). However, much of the above-mentioned research utilized footwear modifications that introduced considerable alterations to the normal functioning of the ankle. The purpose of the current study was to assess the effect of the sagittal CoP position on kinetic and kinematic parameters of the lower extremities. Utilizing a novel foot-worn biomechanical device which allows controlled manipulation of the CoP, we hypothesized that translation of elements in the sagittal plane (i.e., from posterior to anterior) would result in a matching alteration of the magnitude of lower extremity sagittal joint torques and kinematic patterns during the stance phase. Table 1 Demographic data of participants (n=14). Age (years) Height (cm) Weight (kg) Note: Values are mean7sd Table 2 Spatio-temporal parameters, group values (n=14). Parameters Anterior axis Neutral axis Posterior axis Cadence (steps/min) Stride length (m) Walking speed (m/s) Note: Mean values7standard deviation Fig. 1. Biomechanical platform and mobile elements. Notes: The biomechanical device utilized in the study, comprising four modular elements attached onto footworn platforms (APOS system, Apos Medical and Sports Technologies Ltd.). The device consists of two convex-shaped biomechanical elements attached to each of the feet. Each element can be individually calibrated (Position, convexity, height and resilience) to induce specific biomechanical challenges in multiple planes. The elements are available in different degrees of resilience and convexity, and are attached to the subject s foot using a platform in the form of a shoe. Fig. 3. Representative subject s CoP relative offset at the posterior, neutral and anterior configurations. The Y axis represents the vertical distance between the instantaneous location of the CoP of the instantaneous axis heel axis (perpendicular to the heel to axis crossing the heel marker. (All values are reported in mm and negative values indicate lateral offset). The X represents 1% of stance phase time. Fig. 2. A. Biomechanical device at neutral sagittal configuration, B. at anterior configuration, C. at posterior configuration. In the anterior and the posterior configurations the biomechanical elements (red spheres) are transposed anterior and posterior in relation to the neutral configuration conveying matched offset of the COP. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3 A. Haim et al. / Journal of Biomechanics 43 (21) Methods 2.1. Participants Fourteen healthy male volunteers without any known musculoskeletal or neurologic pathology comprised the study cohort. All had the same shoe size (French 43) and a similar anthropometric profile (i.e., weight, height, dominant leg). Their characteristics are noted in Table 1. The study was approved by the Ethics Sub-Committee and all participants gave informed consent The biomechanical system A novel biomechanical device (APOS System, APOS Medical and Sports Technologies Ltd. Herzliya, Israel) allowing controlled manipulation of the CoP was generously donated by the manufacturer prior to the study. A detailed description of the device was recently reported (Haim et al., 28). In brief, it consists of two mobile convex-shaped biomechanical elements attached to each of the feet, enabling flexible continuous positioning in multiple planes (Fig. 1). A pilot study conducted to assess the stability of the apparatus determined that, for healthy adults, satisfactory walking stability can be kept within the range of 1.8 cm posterior and 1.8 cm anterior deviation of the biomechanical elements from the neutral position. Table 3 Comparison of average CoP sagittal trajectory (n=14). Device configuration Anterior Neutral Posterior p Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Stance phase stage Initial contact o.1 Load response o.1 Midstance o.1 Terminal stance o.1 Pre-swing o.1 Terminal contact o.1 Note: Values represent the instantaneous CoP-heel axis vertical distance; values reported in mm 2.3. Experimental protocol Prior to testing, all participants were functionally assessed by the same physician (AH). The biomechanical device was calibrated by a qualified physiotherapist. First positioning of the elements for the functional neutral configuration, defined as the position in which the apparatus exerted the least valgus, varus, dorsal or plantar torque about the ankle to the individual being examined was determined. Anterior and posterior configurations were then defined as 1. cm anterior and 1. cm posterior deviation of the biomechanical elements along the neutral sagittal axis (Fig. 2). Successive gait analysis testing, each with a singular calibration of the apparatus, was performed with the biomechanical elements placed in three conditions: neutral configuration (Fig. 2a), anterior displacement (Fig. 2b), and posterior displacement (Fig. 2c). To become accustomed to the testing procedure, subjects were instructed to walk at a self-selected velocity for several minutes which was then indicated by a metronome to ensure consistent cadence throughout the trial. Six successful trials of each condition were collected per subject for averaging. All conditions were tested in random order on the same day Data acquisition and processing Gait analysis of each subject was performed at the Biorobotics and Biomechanics Lab at Technion-Israel Institute of Technology. Three-dimensional motion analysis was performed using an 8-camera Vicon motion analysis system(oxford Metrics Ltd., Oxford, UK) for kinematic data capture, at a sampling frequency of 12 Hz. The ground reaction forces were recorded by two 3-dimensional AMTI OR6-7-1 force plates placed in tandem in the center of a 1-m walkway, at a sampling frequency of 96 Hz. Kinematic and kinetic data were collected simultaneously while the subjects walked over the walkway. A standard marker set was used to define joint centers and axes of rotation (Kadaba et al., 199). Markers were attached bilaterally over the following anatomic landmarks: the anterosuperior iliac spine, the posteriosuperior iliac spine, the lateral midthigh, the lateral knee epicondyle, the lateral midshank,the lateral malleolus, the head of the third metatarsal, and the posterior aspect of the heel at the same level as the marker over the third metatarsal head. A knee alignment device (KAD; Motion Lab Systems Inc, Baton Rouge LA) was utilized to estimate the threedimensional alignment of the knee flexion axis during the static trial. Sagittal plane joint angles and torques were calculated using inverse dynamic analyses from the kinematic data and force plates measures using PlugInGait (Oxford Metrics, Oxford, UK). All analyses were performed for the dominant leg. Joint moments were normalized for body mass. To examine the relationship between the different interventions on the outcome measures, trial data were extracted and calculated by MATLAB TM software. Stride time normalized curves of the joint angles and moments were plotted. All values were reported in association with a specific stage of the gait cycle: initial contact (IC) 2%; load response (LR) 1%; midstance (MS) 1 3%; terminal stance (TS) 3 %; pre-swing (PS) 6%; terminal contact (TC) 6%- Table 4 Comparison of the average joint kinematic parameters (mean and SD). Anterior Neutral Posterior p Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Knee Total range of motion (throughout gait cycle) Initial contact Peak flexion (midstance) Peak extension (terminal stance) Terminal contact Peak flexion (swing phase) Ankle Total range of motion (throughout gait cycle) Initial contact Peak plantar flexion at loading response Peak dorsal flexion at midstance Terminal contact Hip Total range of motion (throughout gait cycle) Initial contact Peak flexion (loading response) Peak extension (terminal stance) Terminal contact

4 972 A. Haim et al. / Journal of Biomechanics 43 (21) toe off (Perry, 1992). Relative CoP offset and peak values of joints angles and torques during different phases of the stance period were calculated and their average was determined across six trials in each configuration for each subject. The relative duration of the knee flexor torque at MS and extensor torque at TS (torque duration/total gait cycle duration) was calculated as well. The individual values of each subject were used for inter-group statistical analysis. Calculation of the CoP trajectory with instantaneous coordinates of the CoP recorded by the force plate and matching instantaneous coordinates of the heel and toe markers was carried out; this method was recently described by our group (Haim et al., 28). Total CoP offset (i.e., the relative distance of the CoP from the neutral configuration) and the offset at IC, LR, MS, TS, PS and TC stance stages were calculated. 2.. Statistical analysis The null hypothesis that the joint angles and moment s magnitude were the same for each of the walking conditions was tested each of the parameters. Nonparametric Friedman tests were used for comparison of spatio-temporal (cadence, step length, gait velocity), kinetic, kinematic and CoP offset parameters in the neutral, anterior and posterior configurations of the apparatus. For the significant results we further used Wilcoxon tests to compare each pair from the three groups. Spearman s correlations were used to examine the relationship of kinetic parameters in the posterior, neutral and anterior configuration of the apparatus. A probability of less than. was considered as statistically significant. All analyses were performed using SPSS (version 13.). 3. Results 3.1. Temporal spatial variables Cadence and walking velocity were similar for all configurations of the apparatus. The stride length was 3 cm longer for the posterior condition compared to the anterior condition; however, this was not statistically significant (Table 2) CoP trajectory The CoP trajectory throughout stance shifted in accordance to the offset of the biomechanical elements (Fig. 3). Inter-subject analysis revealed a significant relationship between CoP locus throughout stance and the sagittal offset of the biomechanical elements from the neutral position (Table 3) Sagittal plane kinematics There were significant differences in ankle, knee and hip kinematics between the three test conditions (Table 4, Fig. 4a c). Ankle: Sagittal plane ankle total range of motion (RoM) was similar for all conditions tested. At IC, the ankle was slightly dorsiflexed in all conditions tested ( on average). Anterior and posterior offset significantly related with greater and lesser dorsal flexion, respectively, on average, % of total RoM. Immediately after IC, during LR, the ankle planter flexed. Peak plantar flexion was significantly greater in the posterior condition than in the anterior condition, on average, % of total RoM. During MS, the joint dorsal flexed. Peak dorsal flexion at the end of MS was not statistically significant for the three walking conditions. Finally, during PS, the ankle plantar flexed once more. Prior to TC, peak plantar flexion was significantly greater in the anterior condition than in the posterior condition, on average, % of total RoM. Sagittal plane joint kinematics Ankle Knee Hip Joint angle ( ) % Gait Cycle Fig. 4. Sagittal plane joint kinematics. (a c): representative subject s sagittal plane joint kinematics for the three walking conditions tested: neutral (yellow), anterior (red) posterior (green) configurations. The Y axis represents joint angles and the X axis represents 1% of a single gait cycle. Data was sampled at the following: the intersection of the curve with Y axis represents initial foot contact (IC). The vertical lines represent terminal foot contact (TC). a n peak ankle planter flexion at loading response (LR); a nn peak ankle dorsal flexion at terminal stance (TS); a nnn peak ankle planter flexion at pre-swing (PS); b n peak knee flexion at midstance (MS); b nn peak knee extension at terminal-stance (TS); c n peak hip flexion at loading response (LR); c nn peak hip extension at pre-swing (PS). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) 2 1-1

5 A. Haim et al. / Journal of Biomechanics 43 (21) Knee: Sagittal plane knee total RoM was similar in all conditions tested. At IC, knee extension was on average (.8% of total RoM) greater for the posterior condition than for the anterior condition. During LR phase, the knee flexed for the first time. Peak knee flexion (during early MS) was similar for the three walking conditions. Following this flexion peak, the knee extended. Peak knee extension at TS was slightly greater with the posterior shoe configuration (on average, % of total RoM) than in the anterior condition. Finally, during PS, the knee flexed for the second time and knee flexion was on average (11.2% of total RoM) greater at TC in the anterior condition than in the posterior condition. Peak knee flexion occurring during the swing phase was similar for all walking conditions. Hip: At IC, hip flexion was on average (4.8% of total RoM) greater in the posterior condition than in the anterior condition. Peak hip flexion (during the end of LR and early MS) was on average 1.61 (3.72% of total RoM) greater in the posterior condition than in the anterior condition. During mid and TS phase, the hip extended. Peak hip extension (during the end of TS and early PS) was similar in all conditions tested. At TC, the hip flexion was on average (7.37% of total RoM) greater in the anterior condition than in the posterior condition. Table Comparison of average joint kinetic parameters (mean and SD). Anterior Neutral Posterior p Mean Std. Dev. Mean Std. Dev. Mean Std. Dev. Knee Peak extension torque at loading response (N m/kg) Peak flexion torque at midstance (N m/kg) Midstance flexor moment duration (% gait cycle) Peak extension torque at terminal stance (N m/kg) Terminal stance extensor moment duration (% gait cycle) Peak flexion torque at pre-swing (N m/kg) Ankle Initial contact (N m/kg) Peak ankle planter flexion at loading response (N m/kg) Peak ankle dorsal flexion torque at pre-swing (N m/kg) Hip Peak hip flexion torque at loading response (N m/kg) Peak hip extension at pre-swing (N m/kg) Sagittal plane joint kinetics Ankle Knee Hip Joint moment (N-m/k) % Gait Cycle Fig.. Sagittal plane joint kinetics. (a c): representative subject s sagittal plane joint kinetics for the three walking conditions tested: neutral (yellow), anterior (red), posterior (green). The Y axis represents joint angles and the X axis represents 1% of a single gait cycle. The intersection of the curve with Y axis represents initial foot contact (IC). The vertical lines represent terminal foot contact (TC). a n peak ankle planter flexion torque at loading response (LR); a nn peak ankle dorsal flexion torque at pre-swing (PS); b n peak knee extension torque at loading response (LR); b nn peak knee flexion torque at midstance (MS); b nnn peak knee extension torque at terminal stance (TS); b nnnn peak knee flexion torque at pre-swing (PS); c n peak hip flexion torque at loading response (LR); c nn peak hip extension torque at pre-swing (PS). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

6 974 A. Haim et al. / Journal of Biomechanics 43 (21) Sagittal plane kinetics There were significant differences in ankle, knee and hip kinetics between the three test conditions (Table, Fig. a c, Figs. 6 8). Ankle: A significant correlation was found between the device configuration and the ankle torque at LR and at TS (Table 6). At IC (Fig. a), ankle dorsal flexion torque was lower by.41 N m/kg for the posterior configuration, as compared to the anterior configuration, a 63% reduction (Fig. 6, Table ). Following IC, during LR, the reaction force passes behind the joint and generates a plantar flexion torque about the ankle. Peak plantar flexion torque (during LR) was on average 2.17 N m/kg greater for the posterior condition than for the anterior condition, a 77.22% increase. During midstance and TS, the reaction force passes in front of the joint center. The joint sagittal plane external torque is transformed to a dorsal flexion torque. Peak dorsal flexion torque (at the end of TS and the beginning of PS) was 1. N m/kg greater in the anterior condition, a 4.76% rise. Knee: A significant correlation was found between the device configuration and the knee torque throughout the stance phase (Table 6). Immediately after IC, the reaction force passes in front of the knee (Fig. b). On average, the peak torque was 2.1 N m/kg greater for the anterior condition, a 44.42% rise (Fig. 7, Table ). During MS, the line of action passes behind the knee and the torque reverses into a flexion torque. This torque peaks early in MS with the peak flexion angle of the knee. The peak torque was 1.97 N m/kg lower for the anterior condition than for the posterior condition, a 23.12% reduction. During TS, as the center of mass passes the base of support, the reaction force once again passes in front of the knee and the torque reverses into an extension torque. The magnitude of the peak torque was 1.12 N m/kg greater for the anterior condition than for the posterior, a.6% change. In two subjects, the sagittal knee torque remained flexed with the posterior configuration throughout the entire stance period. These subjects were excluded from the analysis of flexor/extensor torque duration. For the remaining 12 subjects extensor torque was significantly longer with the anterior shoe configuration and the flexor torque was shorter, although this difference was not statically significant (Table ). Throughout PS, the reaction force passes just behind the joint center and induces a flexion torque; the peak torque was less for the anterior configuration in comparison to the posterior configuration, a 34% reduction. Hip: A significant correlation was found between the device configuration and the hip torque at MS (Table 6). At IC, the GRF passes in front of the hip, bringing on a flexion sagittal torque. This torque peaks during LR (Fig. c). The magnitude of the torque was similar in the three walking conditions. The torque then diminishes and transforms into an extension torque which peaks during PS. On average, the peak extension moment was 1.28 N m/ kg lower for the anterior configuration compared to the posterior, a 9.8% reduction (Fig. 8, Table ). Sagittal torque (Nm/Kg) Loading response Mid stance Terminal stance Pre wing Sagittal joint angle ( ) Anterior Neutral Posterior Anterior Neutral Posterior Anterior Neutral Posterior Anterior Neutral Posterior Fig. 6. Knee kinetics and kinematics during stance phase stages. Notes: relationship between group joint sagittal moment values throughout consecutive stages of gait cycle and concomitant joint sagittal angles. Data presented as box-plots line in center of box represents the median peak value; the box represents the interquartile range and the whiskers represent the range.

7 A. Haim et al. / Journal of Biomechanics 43 (21) Discussion The results presented indicate a clear association between the magnitude of lower extremity kinetic parameters and the position of the CoP in the sagittal plane. The present study examined the outcome of a controlled shift of the CoP in healthy subjects. Several footwear-generated biomechanical manipulations (e.g., high heels, reverse heel, rocker bottom) have been shown to influence movement patterns in the sagittal plane. However, these interventions introduce vigorous interference to ankle kinematics. To the best of our knowledge, this is the first study to utilize a biomechanical device which allows controlled modulation of the center of pressure. We found that anterior translation of the CoP in the sagittal axis correlated with an ankle dorsal flexor and a knee extension shift of the sagittal torque throughout the stance phase, a reduced extension torque at the hip during PS and a prolonged duration of the terminal stance knee extension torque. A reverse outcome was found with posterior CoP translation. These findings confirm the study s hypothesis of a direct correlation between the sagittal location of the CoP and the magnitude of lower extremity sagittal joint torques. We speculate that the sagittal shifted CoP reduced or extended the distance between the GRF and the center of the joints throughout successive stages of the stance phase, resulting in reduced or increased magnitude of the torques. Kinematic patterns of the ankle, knee and hip joints were also found to be influenced by a sagittal shift of the CoP. Sagittal translation of the CoP from posterior to anterior offset correlated with a flexion shift of the knee kinematic patterns and with a bimodal pattern of the ankle and hip kinematics (ankle plantar flexion/hip extension during initial stance and ankle dorsal flexion/ hip flexion during final stance). Kerrigan et al. (2) examined the effect of high-heeled shoes on gait parameters in healthy women and reported a 2.41 increase in ankle plantar flexion throughout the gait cycle. In the present study, the effect of anterior and posterior CoP translation on ankle kinematics was less profound. Preserving normal ankle function enables a controlled setting for easement of CoP influence on kinetic parameters. Initial-contact Loading response Pre swing Sagittal torque (Nm/Kg) Sagittal joint angle ( ) Anterior Neutral Posterior Anterior Neutral Posterior Anterior Neutral Posterior Fig. 7. Ankle kinetics and kinematics during stance phase stages. Notes: relationship between group joint sagittal moment values throughout consecutive stages of gait cycle and concomitant joint sagittal angles. Data presented as box-plots line in center of box represents the median peak value; the box represents the interquartile range and the whiskers represent the range.

8 976 A. Haim et al. / Journal of Biomechanics 43 (21) Sagittal torque (Nm/Kg) Sagittal joint angle ( ) Loading response 1 Anterior Neutral Posterior Pre swing 1 Anterior Neutral Posterior Fig. 8. Hip kinetics and kinematics during stance phase stages. Notes: relationship between group joint sagittal moment values throughout consecutive stages of gait cycle and concomitant joint sagittal angles. Data presented as box-plots line in center of box represents the median peak value; the box represents the interquartile range and the whiskers represent the range. Kerrigan et al. (2) reported greater peak knee flexion, prolonged knee flexor torque and reduced peak knee-extensor torque with high heels. In the present study, posterior CoP offset correlated with similar kinetic findings (i.e., greater and prolonged knee flexor torque and reduced shorter peak kneeextensor torque). The knee sagittal torque significantly correlated with the knee flexion angle. Interestingly, with posterior offset configuration, knee angles were not significantly different during midstance and were more extended during terminal stance. This suggests that the altered kinetics recorded with the posterior offset is a result of altered position of the GRF and is not caused by altered joint kinematics (i.e., increased knee flexion angles with posterior CoP offset could have accounted for the greater flexor torque). Several limitations arising from the current study should be noted. First, the relative CoP location was analyzed indirectly by calculating instantaneous force plate recorded COP and corresponding foot segment axis distance. While, this method offers a reasonable evaluation of the COP offset and was utilized in previous studies (Haim et al., 28), future studies incorporating direct COP measurement (e.g., pedobarograph analysis) could provide valuable data regarding shoe COP pattern modulation. Another limitation of this study was the employment of the apparatus at neutral position as a control. This setting was selected to assure consistency of the kinematic model. Finally, it should be emphasized that the participants in this study comprised a distinctive homogenic cohort of healthy young male adults. These results are therefore valid only for individuals with characteristics similar to those of the tested group. Different populations (e.g., females who tend to have different lower extremity joint motions compared to males due to anatomical, muscle strengths, ligament properties) may respond differently to such interventions. Further studies are needed before these findings can be validated in other populations. The results of the present study offer clinically relevant implications to several musculoskeletal pathologies. The knee Table 6 Spearman s correlations analysis of kinetic parameters and device configuration. Test variable Device configuration Anterior Neutral Posterior Ankle moment (IC) Anterior 1; (.) Neutral.67; (.9) 1; (.) Posterior.612; (.2).4;(.66) 1; (.) Ankle peak plantar flexion moment (LR) Anterior 1; (.) Neutral.943; (.) 1; (.) Posterior.71; (.).63; (.1) 1; (.) Ankle peak dorsal flexion moment (PS) Anterior 1; (.) Neutral.824; (.) 1; (.) Posterior.783; (.1).99; (.) 1; (.) Knee peak extensor moment (LR) Anterior 1; (.) Neutral.873; (.) 1; (.) Posterior.6; (.22).739; (.3) 1; (.) Knee peak flexor moment (MS) Anterior 1; (.) Neutral.974; (.) 1; (.) Posterior.842; (.).899; (.) 1; (.) Knee peak extensor moment (TS) Anterior 1; (.) Neutral.92; (.) 1; (.) Posterior.93; (.).934; (.) 1; (.) Knee peak flexor moment (PS) Anterior 1; (.) Neutral.912; (.) 1; (.) Posterior.96; (.).96; (.) 1; (.) Hip peak flexor moment (LR) Anterior 1; (.) Neutral.873; (.) 1; (.) Posterior.4 (.11).431 (.124) 1; (.) Hip peak extensor moment (MS) Anterior 1; (.) Neutral.982; (.) 1; (.) Posterior.91; (.).924; (.) 1; (.) Values are correlation coefficients (r), P values in parentheses. Abbreviations: IC Initial contact; LR Loading response; PS pre-swing; MS Midstance; TS terminal stance

9 A. Haim et al. / Journal of Biomechanics 43 (21) flexion moment during MS is proportional to the pressure across the patellofemoral joint and has been linked with patellofemoral pain syndrome (PFPS) and osteoarthritis (OA) of the knee (Kerrigan et al., 1998). Similarly, it has been suggested (Astephen et al., 28) that interventions designed at altering knee kinetics may be effective for halting progression of knee OA. Secondly, in anterior curciate ligament (ACL)-deficient knees, internal moment generated by quadriceps contraction can cause excessive anterior tibial translation. It has been suggested that this motion can lead to premature knee osteoarthritis. A reduction in the peak knee flexion moment coupled by a reduced internal quadriceps moment has been reported to be a necessary compensation to avoid excessive anterior translation of the tibia (Andriacchi and Dyrby, 2). Finally, patients suffering from cerebral palsy and other neurological pathologies often experience difficulty maintaining upright posture due to a reduction in the total support moment (Lampe et al., 24). Biomechanical manipulation via A footwear design that incorporates anterior CoP offset may induce an extension shift to the sagittal torque and provide benefit to these patients. An extension shift to the sagittal torque could theoretically lower patellofemoral joint pressure in knee OA patients, diminish excessive anterior tibial translation in patients with ACL deficient knees, and contribute to total support moment in patients with cerebral palsy. However, such interventions should be taken with caution; excessive extension shift to the sagittal torque could possibly alter joint kinematics. it should be mentioned that an extension shift to the sagittal torque may not be safe for the knee. A reduced tendency to flex the knee can reduce the knee joint s capacity for shock absorption and would likely aggravate the tibiofemoral contact stresses at the articular cartilage. Further studies examining the benefit and safety of moderate anterior CoP offset alterations in patients with the above pathologies are warranted. Conflict of interest statement No author has any conflict of interest to declare. Acknowledgment The authors thank APOS Medical and Sports Technologies Ltd. for their generosity in contributing the devices used in the study. References Andriacchi, T.P., Dyrby, C.O., 2. Interactions between kinematics and loading during walking for the normal and ACL deficient knee. Journal of Biomechanics 38, Astephen, J.L., Deluzio, K.J., Caldwell, G.E., Dunbar, M.J., Hubley-Kozey, C.L., 28. Gait and neuromuscular pattern changes are associated with differences in knee osteoarthritis severity levels. Journal of Biomechanics 41, Attinger-Benz, D., Stacoff, A., Balmer, E., Durrer, A., Stuessi, E., Walking pattern with missing heel shoes. In: Proceedings of 11th Conference of the ESB (European Society of Biomechanics), Toulouse, France, pp Crenshaw, S.J., Pollo, F.E., Calton, E.F., 2. Effects of lateral-wedged insoles on kinetics at the knee. Clinical Orthopaedics and Related Research 37, de Lateur, B.J., Giaconi, R.M., Questad, K., Ko, M., Lehmann, J.F., Footwear and posture. Compensatory strategies for heel height. American Journal of Physical Medicine and Rehabilitation 7, Gronley, J.K., Perry, J., Gait analysis techniques. Rancho Los Amigos Hospital Gait Laboratory. Physical Therapy 64, Haim, A., Rozen, N., Dekel, S., Halperin, N., Wolf, A., 28. Control of knee coronal plane moment via modulation of center of pressure: a prospective gait analysis study. Journal of Biomechanics 41, Kadaba, M.P., Ramakrishnan, H.K., Wootten, M.E., 199. Measurement of lower extremity kinematics during level walking. Journal of Orthopaedic Research 8, Kakihana, W., Akai, M., Nakazawa, K., Takashima, T., Naito, K., Torii, S., 2. Effects of laterally wedged insoles on knee and subtalar joint moments. Archives of Physical Medicine and Rehabilitation 86, Kerrigan, D.C., Todd, M.K., Riley, P.O., Knee osteoarthritis and high-heeled shoes. Lancet 31, Kerrigan, D.C., Johansson, J.L., Bryant, M.G., Boxer, J.A., Della Croce, U., Riley, P.O., 2. Moderate-heeled shoes and knee joint torques relevant to the development and progression of knee osteoarthritis. Archives of Physical Medicine and Rehabilitation 86, Lampe, R., Mitternacht, J., Schrodl, S., Gerdesmeyer, L., Natrath, M., Gradinger, R., 24. Influence of orthopaedic-technical aid on the kinematics and kinetcs of the knee joint or patients with neuro-orthopaedic disease. Brain and Development 26, Maly, M.R., Culham, E.G., Costigan, P.A., 22. Static and dynamic biomechanics of foot orthoses in people with medial compartment knee osteoarthritis. Clinical Biomechanics (Bristol, Avon) 17, Myers, K.A., Long, J.T., Klein, J.P., Wertsch, J.J., Janisse, D., Harris, G.F., 26. Biomechanical implications of the negative hell rocker sole shoe: gait kinematics and kinetics. Gait and Posture 24, Ogata, K., Yasunaga, M., Nomiyama, H., The effect of wedged insoles on the thrust of osteoarthritic knee. International Orthopaedics 21, Perry, J., In: Gait Analysis: Normal and Pathological Function. Slack, Inc., Thorofare, NJ. Snow, R.E., Williams, K.R., High heeled shoes: their effect on center of mass position, posture, three-dimensional kinematics, rearfoot motion, and ground reaction forces. Archives of Physical Medicine and Rehabilitation 7, Winter, D., Kinematic and kinetic patterns in human gait. Human Movement Science 3, Xu, H., Akai, M., Kakurai, S., Yokota, K., Kaneko, H., Effect of shoe modifications on center of pressure and in-shoe plantar pressures. American Journal of Physical Medicine and Rehabilitation 6, Yasuda, K., Sasaki, T., The mechanics of treatment of the osteoarthritic knee with a wedged insole. Clinical Orthopaedics and Related Research 21,